电气工程学报 ›› 2020, Vol. 15 ›› Issue (2): 24-33.doi: 10.11985/2020.02.004

• • 上一篇    下一篇

扫码分享

虚拟电厂参与大规模新能源系统的储能侧双层优化运行策略*

王开科1(),南东亮1,2(),李勇3,张路1,李笑竹2   

  1. 1.国网新疆电力有限公司电力科学研究院 乌鲁木齐 830013;
    2.新疆大学可再生能源发电与并网技术教育部工程研究中心 乌鲁木齐 830047;
    3.国家电网公司华中分部 武汉 430008
  • 收稿日期:2020-02-03 出版日期:2020-06-25 发布日期:2020-07-31
  • 作者简介:王开科,男,1968 年生,高级工程师。主要研究方向为电力系统安全稳定控制。E-mail:wangkaike@dky.xj.sgcc.com.cn|南东亮,男,1985 年生,硕士,高级工程师。主要研究方向为可再生能源发电与并网技术。E-mail:nandongliang@dky.xj.sgcc.com.cn
  • 基金资助:
    * 国家自然科学基金(51667020);新疆维吾尔族自治区高校科研计划自然科学项目(XJEDU2017I002);国网新疆电力有限公司科技项目(5230DK19000B)

Bi-level Optimal Operation Strategy of Energy Storage with Large-scale New Energy System under the Participation of Virtual Power Plant

WANG Kaike1(),NAN Dongliang1,2(),LI Yong3,ZHANG Lu1,LI Xiaozhu2   

  1. 1. Electric Power Research Institute of State Grid Xinjiang Electric Power Co., Ltd., Urumqi 830013 China;
    2. Engineering Research Center of Ministry of Education for Renewable Energy Generation and Grid Connection Technology, Xinjiang University, Urumqi 830047 China;
    3. Central China Branch of State Grid Corporation, Wuhan 430008 China
  • Received:2020-02-03 Online:2020-06-25 Published:2020-07-31

摘要:

为响应大规模新能源系统储能侧的灵活运行调度,提出一种虚拟电厂(Virtual power plant,VPP)参与储能系统的双层优化运行模型。模型中利用威布尔、贝塔、耿贝尔分布描述风电、光伏、水电的发电不确定性,计及新能源出力不确定性的高估低估成本;计及VPP以提高系统灵活性,模型中将VPP作为独立市场主体参与系统调度,两层之间交替求解,实现协同优化。针对模型混合整数、非凸、非线性、多目标的特点,提出一种新型的多种群伪并行多目标算法并进行求解,利用修改后的IEEE39节点为算例进行验证,并提出此模型可均衡各层市场主体,在保证系统安全运行的基础上运营利润最大,有效实现电力系统的可持续性发展,同时也验证了多种群伪并行多目标算法的有效性与竞争性。

关键词: 大规模新能源, 虚拟电厂, 储能, 双层优化, 多目标算法

Abstract:

In response to the flexible operation scheduling of the energy storage of large-scale new energy system, a bi-level optimal model with the participation of virtual power plant (VPP) is proposed. Weibull, Beta and Gumble distribution are used to describe the uncertainty of wind, PV and hydropower, and the overestimation and underestimation cost of new energy output are considered. Introducing VPP to improve the flexibility, the model takes VPP as an independent market subject to participate in scheduling, and solves the problem alternately between the two levels to achieve collaborative optimization. According to the characteristics of mixed integer, nonconvex, nonlinear and multi-objective, a novel multi group pseudo parallel multi-objective algorithm is proposed to solve the problem. The modified IEEE39 bus system is used as an example to verify the model, it can balance the market entities at all levels, ensure the safe operation of the system and maximize the operating profit, and effectively realize the sustainable development of the power system. The effectiveness and competitiveness of the algorithm are verified simultaneously.

Key words: Large-scale new energy, virtual power plant, energy storage, bi-level optimization, multi-objective algorithm

中图分类号: